Wetlands are important ecosystems formed by an interaction between water and land, and play a vitally important role in regulating the water cycle, reducing greenhouse gas emissions, breaking down pollution, and protecting biodiversity (Chen et al., 2018; Zhang et al., 2019). The Sanjiang Plain wetland is the largest freshwater wetland distribution area in northeastern China, and is essential for protecting biodiversity and maintaining an ecological balance (Zhao, 1999; Dong et al., 2022). In recent years, the scale of irrigation has expanded due to the need for agricultural development, and groundwater levels have subsequently experienced an annual decrease (Chen et al., 2018; Qu et al., 2022). Wetland water levels have been lowering continuously, and the wetlands themselves have in turn changed from swamps to swampy-, typical-, or miscellaneous grass meadows (Dong et al., 2022; Luo et al., 2022). As a result, these wetland ecosystems have become spatially heterogeneous. In ecology, spatial heterogeneity is characterized by ecosystem patchiness and changes in environmental gradients (Dong, 2011). Ecosystem patchiness shows that resource levels within patches are relatively consistent; however, large differences exist between patches (Wijesinghe and Handel, 1994). The widespread existence of environmental heterogeneity increases the difficulty in which plants obtain and utilize essential resources (Benton et al., 2003; Chu et al., 2020; Luo et al., 2022). The uptake and utilization of necessary resources for plant growth are linked to environmental suitability, which thus enables plants to develop adaptive strategies for efficiently accessing the necessary resources (Benton et al., 2003; Dong, 2011).

Plants have developed adaptive strategies toward their environments to cope with inter-and intraspecific competition during all their life stages (Sugiyama and Bazzaz, 1998; Pierce et al., 2013, 2017). The study of reproduction has been at the heart of evolutionary ecology (Hodgson et al., 1999; Rüger et al., 2018). The CSR theory, proposed by Grime (1974), greatly promoted the development of a new stage of life-history strategy research, namely from qualitative description to quantitative analysis. A plant life history strategy is the most suitable trade-off strategy between the environment and plant characteristics that gradually formed over long-term evolution (Grime, 2001; Caccianiga et al., 2006). Both environmental and genetic traits influence plant life-history strategies. According to plant productivity levels and the stress intensity produced by the environment, CSR theory classifies plant ecological strategies into three types: C-competitors, S-tolerators, and R-ruderals. C-competitors generally have good competitiveness and plasticity, and their energy utilization is mainly directed toward monopolizing resources quickly in fertile habitats to obtain the highest population competitiveness (Grime, 1974; Dayrell et al., 2018). S-tolerators, which have low population growth rates, short statures, and infrequent flowering events, invest energy into ecological adaptation behaviors, such as stress resistance (Reich, 2014; Dayrell et al., 2018). R-ruderals have a good colonization capacity, and are characterized by rapid germination and flowering rates, which allow them to quickly invest resources and energy into reproduction (Pierce et al., 2017; Dayrell et al., 2018). After years of development and validation, CSR strategies have been heavily supplemented. Hodgson et al. (1999) provide a method for calibrating the classification of CSR strategies: (1) the degree of species dominance in productive and undisturbed habitats provides a C-selection index, (2) variability in leaf traits known to reflect conservative lifestyles provides an S-selection index, and (3) species frequency in potentially productive but disturbed habitats provides an R-selection index (Hodgson et al., 1999). Pierce et al. (2013) used specific quantitative traits for analysis, and applied the method to general vascular plants, while simultaneously adapting it to the comparison of species within a colony and individuals within a population. Based on functional ecology theory, principal component analysis (PCA) is used to transform functional trait variation into a triangular order corresponding to CSR theory, which has a theoretical and practical application value (Pierce et al., 2013).

Reproduction is an important part of a plant’s life history. Not only is it the basis of population formation, development, and evolution, but also of the condition of ecosystem evolution (Chen and Wang, 2014; Rüger et al., 2018; Chen et al., 2019). Different plants use different reproduction methods to maintain and expand their populations (Jackson et al., 1985). Resource allocation theory holds that the biomass fixed by photosynthesis is quantitative, and that plants cannot invest an equal amount of biomass into each trait (Cornwell et al., 2014). Thus, trade-offs must be made between traits for plants to adapt to the environment (Jackson et al., 1985). Increased resource allocation to one trait may therefore decrease resource allocation to another trait (Hodgson et al., 1999). For example, a trade-off exists between clonal growth and sexual reproduction such that an increased investment in sexual reproduction will reduce the investment in clonal reproduction (Levins, 1968). This trade-off may be determined by environmental conditions, plant size, population density, and heredity (Abrahamson, 1975; Reekie, 1991). Díaz posited that the trade-off among functional traits determines plant ecological countermeasures (Díaz et al., 2016), and that ecological countermeasures are the basic conditions required to adapt to habitat diversity. The study of plant ecological responses has long been a vital element in plant ecology (Pierce et al., 2012). Only through the rational coordination of resource allocation among traits can plants achieve optimal allocation in complex ecosystems, which provides a competitive advantage for survival and reproduction (Pierce et al., 2017).

Deyeuxia angustifolia is a typical perennial, rhizomatous Poaceae species, and is widely distributed in northeastern China, the Russian Far East, Mongolia, Korea, and Japan. In China, D. angustifolia is the most concentrated in the Sanjiang Plain (Ma, 1995; An et al., 2018; Ni et al., 2022). Under suitable environmental conditions, D. angustifolia mainly relies on vegetative reproduction. This produces a large area consisting of a single, dominant community, and sexual reproduction only occurs to produce offspring. In the past, D. angustifolia research mainly focused on physiological and ecological characteristics, grassland productivity, and the effects of nitrogen deposition (An et al., 2018; Song et al., 2019; Wang et al., 2019). However, a few studies focused on D. angustifolia population reproduction in heterogeneous habitats, and more specifically on plant sexual-and asexual reproduction characteristics, as well as reproduction strategies. Four wetland types were studied: swamp, swampy meadow, typical meadow, and miscellaneous grass meadow. These were selected to analyze the reproductive characteristics and trade-offs of D. angustifolia, and to explore the following: (1) whether D. angustifolia reproductive characteristics significantly change in heterogeneous habitats; (2) whether D. angustifolia ecological strategies change in heterogeneous habitats; and (3) which environmental factors dominate the changes in the reproductive characteristics of D. angustifolia. This study reveals D. angustifolia ecological strategy transformations in heterogeneous habitats, and provides references and a scientific basis for wetland restoration.

Deyeuxia angustifolia is a clonal, rhizomatous plant type with well-developed underground rhizomes (Shi et al., 2015; Dong et al., 2022). At the end of each growing season, energy is directed toward rhizome growth, and rhizomes produce terminal-and node buds which develop into new plants (Hutchings and Wijesinghe, 1997; Dong et al., 2022). Rhizomes occupy a specific niche space by elongating away from the mother plant (Dong, 2011). In addition, tiller node buds can also develop into new plants. Therefore, rhizome bud growth is closely related to outward population expansion, while tiller node buds are related to the population’s ability to colonize new areas (Chu et al., 2020). Plants develop a variety of life history patterns during environmental adaptation. Specifically, out of all four habitats, the water level of SW was too high, which seriously restricted D. angustifolia growth, consequently leading to aboveground biomass and plant density being the lowest (Tables 2, 3). The water level of SM was slightly lower, and flooding stress therefore reduced. In this case, biomass per unit area, population density, and all the various bud densities were maximized. Rhizome strength was the highest in MG and the lowest in SM (Figure 3). Wetland clone plant rhizomes are mainly used for population expansion. Thus, stronger rhizomes give D. angustifolia an enhanced ability to expand its population (Dong, 2011; Shi et al., 2015; Chu et al., 2020). From SM to TM to MG, D. angustifolia density and biomass gradually decreased. When D. angustifolia occurred in TM and MG under drought stress, or in SW under flooding stress, the density of all bud types decreased, while the proportion of rhizome node buds, the number of rhizome buds, rhizome length, and rhizome biomass per plant increased. Whether at population or individual level, D. angustifolia showed a guerrilla-type clonal configuration in adverse environments. On the one hand, an increased roots biomass increased access to resources and therefore enhanced population competitiveness (Hutchings and Wijesinghe, 1997). On the other hand, rhizomes enabled an escaped from unfavorable habitats (Wijesinghe and Handel, 1994). This result is consistent with Chen et al. (2014) for growth and reproductive responses of two water-holding plants to different water level gradients. Therefore, in a restricted environment, D. angustifolia will increase investment in growth and clonal reproduction, thereby reducing the number of individuals per unit area to adapt to environmental stress (Mclntyre et al., 2009; Dong et al., 2022). This enables it to adapt to a wide range of water gradients.

Life history theory predicts a trade-off between resource allocation among different functional traits (Pierce et al., 2012, 2013; Rüger et al., 2018). This means that, when the total amount of resources is certain, increasing resource allocation to one trait will inevitably reduce resource allocation to other traits (Zhang, 2004; Mclntyre et al., 2009; Dong et al., 2022). Due to competition for nutrient elements, photosynthetic nutrients, and other resources between clonal growth and sexual reproduction, an increased investment in sexual reproduction may weaken the investment in clonal reproduction, and vice versa (Dong, 2011; Shi et al., 2015; Lu et al., 2022). In this study, a significant positive correlation existed between sexual reproductive allocation and rhizome biomass per plant, as well as a significant negative correlation between sexual reproductive allocation and tiller node bud number (Figure 4). In terms of reproductive strategy, an investment in sexual reproduction increases with increased environmental stress, while the investment in asexual reproduction (rhizomes) also increased. In contrast, the investment toward producing tillers and sprouts to form new plants reduced. In a more suitable habitat, D. angustifolia increased tiller node output to reduce sexual reproduction input, increased the tillers formed by tiller node bud germination, reduced the production of rhizomes and rhizome node buds, and reduced the investment in sexual reproduction (Figure 2). We can thus classify D. angustifolia functional traits into two categories: 1) outgoing rhizomes, rhizome buds, and sexual reproductive characteristics, and 2) in-situ growing tillering node buds and population density (Chu et al., 2020). D. angustifolia tended toward an “outgoing guerrilla” clonal configuration in stressed environments at both population and individual levels, and thereby produced an escape mechanism. In suitable habitats, D. angustifolia tended toward an in-situ growing “phalanx” clonal configuration (Jung et al., 2010; Zhang et al., 2022). This is different from Li et al. (2018) on the decrease in sexual reproduction and increase in asexual reproduction of Vallisneria spinulosa in stressful environments. The main reason is that different plants have different adaptation strategies to environmental changes.

Plant functional traits result from adaptation to different environments in the long-term evolutionary process, and can objectively express plant adaptability toward external environments (Mclntyre et al., 2009; Jung et al., 2010). In this study, the ecological strategies of D. angustifolia and its changing functional characteristic in the four habitats correspond to the CSR strategy proposed by British ecologist Grime (1974). D. angustifolia growing in MG and TM corresponded to R-strategies. It allocated resources to reproduction and population expansion mainly by devoting more resources to sexual reproduction and prolonging rhizomes to produce nodal buds (Munoz et al., 2016). It was therefore inclined toward the “Ruderal” adaptation strategy, characterized by a good expanding ability and large investment in reproduction (Grime, 1974). This is similar to Cerabolini et al. in potentially fertile, but disturbed, habitats, where many plants tended toward the R-strategy (Cerabolini et al., 2010). In MG and TM, the investment in sexual reproduction increased (2.07 ± 0.52%; 1.01 ± 0.15%), and rhizomes from asexual reproduction tended to guerrilla clonal growth (23.8 ± 1.5%; 19.6 ± 1.4%) so that plants could quickly expand from stressed sites to more suitable habitats (Pierce et al., 2013). D. angustifolia growing in SM corresponded to the C-strategy, in which it invested more resources into vegetative growth; that is, by increasing tiller nodes buds and ramets to occupy most of the available resources so that resource utilization could be optimized. The C-strategy dominated in favorable habitats and enabled the maintenance of a relatively high rate of resource acquisition in crowded, fertile, and undisturbed habitats (Grime, 2001; Cerabolini et al., 2010). The progeny formed by increasing tiller node bud germination, produced by clonal reproduction, increased resource utilization efficiency. The clonal configuration tended to the “Competitor” type. D. angustifolia therefore corresponded to the S-strategy in SW habitats. Due to the serious stress caused by excessive flooding, most of its resources were spent on maintaining survival, which was mainly reflected in short statured plants with few branches, a low flowering and seed setting rate, and other adaptive characteristics. Most of the previous studies analyzed the functional traits of multiple species and classified plants into various CSR strategies (Caccianiga et al., 2006; Pierce et al., 2013). In this study, we analyzed the functional traits of D. angustifolia under fertile, disturbed, and stressed habitats, and found that the adaptation strategies of D. angustifolia populations in these different habitats were consistent with the three CSR strategy types.

In heterogeneous habitats, plants can maximize access to resources by adjusting their resource allocation patterns, thus enabling them to have a wider ecological range and higher tolerance (Wijesinghe and Handel, 1994). Plant organs (roots, stems, leaves, flowers, and fruits) are formed by meristem differentiation (Cornwell et al., 2014; Johnson et al., 2016), which exhibits a high level of plasticity. Organ composition changes dramatically when affected by the external environment. Plant resource allocation varies with the environment, and functional traits are prioritized differently to maximize adaptability to different environments (Johnson et al., 2016). The main problem affecting performance during wetland degradation is the lack of adequate soil water, which affects soil nutrient conditions (Dong et al., 2022). By applying the Mantel test to reproductive characteristics and soil environmental factors, it was found that soil DON, DOC, AN, NN, WC, and AP were important factors driving the differences in reproductive characteristics between heterogeneous habitats in the Sanjiang Plain. In this study, WC was the most important limiting factor, and WC was negatively correlated with rhizome biomass, sexual plant biomass, and inflorescence biomass. This indicated that, when WC decreases, D. angustifolia increases its investment in rhizome and sexual reproduction. On the one hand, water absorption can be enhanced by increasing the number of rhizome fibrous roots, and on the other hand, the investment in rhizomes and seeds can be increased to avoid environmental stress. When WC decreased, and AN and AP increased, D. angustifolia tended toward the R-strategy. Some studies have shown that increased nitrogen levels promote plant sexual reproduction (Hu et al., 2012; Chen et al., 2019). In this study, AN and DON promoted the biomass of sexual plants, while increased WC reduced this effect (Li et al., 2018; Dong et al., 2022). NN and DOC were significantly and positively correlated with plant density, and when WC, DON, and NN increased simultaneously, D. angustifolia tended toward the C-strategy. This result differed from Tomlinson and O'connor (2004). This may be because, in an environment with sufficient soil nutrients, the reproduction strategy of D. angustifolia tended to increase the number of plants per unit area through the germination of tiller node buds instead of escaping by increasing rhizome investment (Wang et al., 2019; Dong et al., 2022). DOC can be quickly transformed into other components in the soil to provide nutrition for plants. In the habitats with a relatively high soil WC, such as SM, this can promote litter decomposition and increase the content of soluble organic carbon in the soil, which in turn promotes D. angustifolia growth (Chen et al., 2019).

This study focused on the process of habitat change along the SM-TM-MG gradient of heterogeneous habitats. Specifically, D. angustifolia produced an escape mechanism from unfavorable environments by increasing individual plant biomass, sexual reproductive investment, and rhizome input, and reducing biomass per unit area, tiller node buds, and plant density. The quantitative variation in reproductive characters, and the trade-off relationship between them, reflected the adaptation strategy of D. angustifolia to changes in environmental factors in heterogeneous habitats. In MG and TM, D. angustifolia invested more in sexual reproduction and rhizome extension, and tended toward the R-adaptation strategy. Clonal configuration tended to be guerrilla clonal configuration. In SM, D. angustifolia invested more in vegetative growth, and tended to the C-adaptation strategy and intensive clonal configuration. D. angustifolia resisted the flooded environment by maintaining a balance of resources, and tended toward the S-adaptation strategy. The reproductive strategy of D. angustifolia in heterogeneous habitats conforms to CSR theory. Among all the environmental factors, WC had the most significant effect on the reproductive characteristics of D. angustifolia. In the arid MG environment, it maintained a high uptake of AN and DON, and increased plant biomass to maintain its adaptation to environmental drought stress. In SM, which had enough water, it increased the number of individuals per unit area to maximize the use of DOC and other nutrients, and to maintain maximum productivity. This study provides a scientific explanation for understanding the ecological adaptation strategies of wetland plants in heterogeneous habitats.

The original contributions presented in the study are included in the article/supplementary materials, further inquiries can be directed to the corresponding author.

HD, HC, JW, YL and HN designed the study. HD, DH, QY and HZ collected the data. HD, LX, YZ and XF analyzed the data. HD, CL, JW and JX lead the writing with all co-authors. All authors gave final approval for publication.

This research was supported by the Youth Innovation Foundation of Heilongjiang Academy of Science (No. CXJQ2022ZR01), the Central Government Guides Local Science and Technology Development Special Projects (No. ZY20B15), the Heilongjiang Academy of Sciences Double-mention Wild Goose Array Special Program Project-Leading Talents (STYZ2023ZR01), and the Applied Technology Research and Development Program of Heilongjiang Province (GA19C006-5).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.